**** BEGIN LOGGING AT Fri Sep 16 02:59:56 2011 Sep 16 07:14:50 hi all Sep 16 07:36:48 hello lumag Sep 16 07:59:21 hi ant_work, bluelightning Sep 16 08:00:01 hi lumag, ant_work, all Sep 16 08:17:54 so, should we revert the split I did when linux.inc was seemingly deprecated? Sep 16 08:17:56 http://cgit.openembedded.org/cgit.cgi/openembedded/commit/recipes/linux/linux-kexecboot.inc?h=org.openembedded.dev&id=6d94ff134f59c2b76216a2127ea494a8c5816024 Sep 16 08:18:37 I'm doing some further cleanings (e.g. moving logo stuff in linux-2.6.39.inc) Sep 16 08:19:02 btw Sep 16 08:19:05 morning :) Sep 16 08:19:20 do we have MIPS support in meta-handhelds? Sep 16 08:19:43 * Jay7 is thinking about Ben NanoNote again :) Sep 16 08:20:38 btw, jlime people will have hard times soon.. they are still using OE classic and known almost nothing about oe-core.. Sep 16 08:21:05 I fear JLime distro may dead.. Sep 16 08:21:55 what is lates kernel? Sep 16 08:24:10 can't say Sep 16 08:24:26 I still don't know how to get access to jlime repo Sep 16 08:25:21 in oe-classic it was 2.6.37 iirc Sep 16 08:44:44 Jay7: it could be that nanonote deserves its own bsp layer Sep 16 08:45:13 it's a little different from the other devices we have in meta-hh Sep 16 08:46:41 we should really try to reach out to jlime... I have a jornada so maybe I should do that... Sep 16 08:48:53 bluelightning: jlime need you :) Sep 16 08:49:11 but placing ben-nn in own bsp is good idea Sep 16 08:49:29 or it may share bsp with zipit and jornadas Sep 16 08:49:52 well we have some jornada stuff in meta-hh Sep 16 08:51:02 iirc, kristoffer is right person to contact on #jlime Sep 16 08:51:12 and B_Lizzard may be Sep 16 08:52:16 ok I will talk to them Sep 16 08:52:19 * Jay7 may have local expo here in about a month Sep 16 08:52:31 good idea to have all my devices up and usable :) Sep 16 08:52:47 ant_work, why do you want to move logo to version-specific inc? Sep 16 08:52:51 not sure how it's real to reach :) Sep 16 08:53:54 * bluelightning -> office Sep 16 08:57:27 lumag: I would add the logo to SRC_URI, one line before defconfig Sep 16 08:58:01 and the LOGO_SIZE ?= "." could also go there Sep 16 08:58:32 but hen I have to swap the order of the includes Sep 16 08:58:36 but why not just linux.inc? Sep 16 08:58:51 not linux-a.b.c.inc Sep 16 08:59:28 ant_work, I'd judge we should keep .inc files machine-clean. Sep 16 08:59:56 What about moving LOGO_SIZE (maybe as BOOTUP_LOGO_SIZE to lower the risk of collisions) to machine .conf? Sep 16 09:00:24 we have it in zaurus.inc Sep 16 09:01:06 it is here for the 'unset' case Sep 16 09:01:54 in fact, as I see it, we always want a bootlogo in the normal case Sep 16 09:02:09 the thing is people committing new recipes often redeclare SRC_URI = in the linux-latest.bb so logo can be missing Sep 16 09:02:36 I would remove the _append in linux-2.6.39.bb too , ofc Sep 16 09:02:59 (in fact I did all above but have still to test the images) Sep 16 09:03:39 final cherry, I think we can remov trhe DEFAULTY_PREFERENCES block and add COMPATIBLE_MACHINE Sep 16 09:04:16 this in the 'common' inc, actually linux-2.6.39.inc Sep 16 09:05:02 lumag: anyway linux-kexecboot.inc should be refactored as you did with linux.inc Sep 16 09:05:18 probably we can merge back the two again Sep 16 09:05:25 ant_work, yes, most probably. Maybe most patches can be applied as is. I just didn't have time :( Sep 16 09:06:21 at the moment we have 3 inc files, which seems a bit overkill Sep 16 09:42:48 Damn. Sep 16 09:43:12 Missing or unbuildable dependency chain was: ['libopie2', 'bluez4', 'gst-plugins-base', 'virtual/libx11'] Sep 16 10:46:16 lumag: btw have you tested meta-opie with the latest changes? Sep 16 10:46:50 it's still missing opiebluetoothd but I have a basic recipe for that locally Sep 16 10:47:07 bluelightning, I'm currently stuck with making opie compatible with sysroot containing qt4 binaries. Sep 16 10:47:24 Do you plan to use qt3e somewhere? Sep 16 10:47:37 lumag: no, AFAIK it was never really very good Sep 16 10:47:55 My current plan is to make uicmoc install moc2 & uic2 and make qmake.bbclass call uic2 and moc2 Sep 16 10:48:13 ok Sep 16 10:48:36 I have to admit I had never tried to mix qt2 & qt4 in the same rootfs Sep 16 10:48:42 or rather, sysroot Sep 16 10:48:58 lumag: btw is it OK if I submit a patch for the subpath issue? Sep 16 10:49:07 (bitbake) Sep 16 10:49:23 yes, you are welcome. I've tried submitting it but seems it got ignored. Sep 16 10:49:35 really? I did not see that... Sep 16 10:51:00 lumag: did you send it to bitbake-devel@lists.openembedded.org or elsewhere Sep 16 10:51:02 ? Sep 16 10:51:11 yes, probably. I don't remember Sep 16 10:52:42 Damn. Got a strange unpack error with openssl-0.9.8m Sep 16 10:52:47 *0.9.8r Sep 16 11:27:23 bluelightning: I was reading "Prevent IMAGE_FEATURES from contaminating initrd" Sep 16 11:27:45 we have indeed some cruft in our initramfs Sep 16 11:27:58 inexplicably some modules stuff leaking in Sep 16 11:28:19 so let's start adding IMAGE_FEATURES = "" Sep 16 11:28:34 ant_work: kind of breaks the debugging thing though right? Sep 16 11:28:50 but we've stopped using that though I think, right? Sep 16 11:29:25 for the custom-purpose cpio containing kexecboot and kexec we don't really need any cruft Sep 16 11:30:29 imho one may want to debug single binaries, nopt any Sep 16 16:35:21 bbl Sep 16 21:42:44 bluelightning: so we have to revive keymaps and formfactor? Sep 16 22:35:44 ant__: yes Sep 16 22:35:59 ant__: I will bring them in this weekend Sep 16 22:39:33 hmm I still have some patches to integrate also Sep 16 22:40:27 great, thx Sep 16 22:43:05 I'll send a reminder for removing tzcode/tzdata from meta-oe Sep 16 22:47:50 you know the issue happening btw layers, masking some new PV... Sep 16 22:48:46 ant__: it's a problem, but unfortunately it's by design... Sep 16 22:52:00 let's call it (mis)feature Sep 16 22:57:03 ant__: I agree, I'm not particularly keen on it myself Sep 16 22:59:53 mail sent Sep 16 23:02:34 so, core-image-minimal doesn't boot on poodle, last line on screen being Sep 16 23:02:53 "No package system found" Sep 16 23:03:23 * ant__ branches serial cable Sep 16 23:05:24 hmm :( Sep 16 23:05:41 what's complaining about that I wonder... Sep 16 23:06:08 it is deliberate for core-image-minimal to have no package management Sep 16 23:07:49 and I suspect some postprocessing removes kernel and modules as well... Sep 16 23:10:00 ouch :/ Sep 16 23:10:02 http://paste.debian.net/130502/ Sep 16 23:33:15 gn **** ENDING LOGGING AT Sat Sep 17 02:59:57 2011